Anyone find any edge with activity based charts over time based charts? - futures io
futures io



Anyone find any edge with activity based charts over time based charts?


Discussion in Traders Hideout

Updated
      Top Posters
    1. looks_one LaissezFaire with 11 posts (5 thanks)
    2. looks_two MiniP with 7 posts (4 thanks)
    3. looks_3 iantg with 5 posts (16 thanks)
    4. looks_4 cory with 2 posts (5 thanks)
      Best Posters
    1. looks_one iantg with 3.2 thanks per post
    2. looks_two cory with 2.5 thanks per post
    3. looks_3 MiniP with 0.6 thanks per post
    4. looks_4 LaissezFaire with 0.5 thanks per post
    1. trending_up 4,132 views
    2. thumb_up 35 thanks given
    3. group 11 followers
    1. forum 31 posts
    2. attach_file 4 attachments




Welcome to futures io: the largest futures trading community on the planet, with well over 125,000 members
  • Genuine reviews from real traders, not fake reviews from stealth vendors
  • Quality education from leading professional traders
  • We are a friendly, helpful, and positive community
  • We do not tolerate rude behavior, trolling, or vendors advertising in posts
  • We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community.  It's free and simple.

-- Big Mike, Site Administrator

(If you already have an account, login at the top of the page)

 
Search this Thread
 

Anyone find any edge with activity based charts over time based charts?

(login for full post details)
  #1 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received

Hi all,

My main chart for day trading (ES) is the 1-minute chart. I find that most of the time this chart gives a good summary of the price action without too much noise. At times of heightened volatility as seen recently, it can be too blunt. A solution during these periods of time may simply be to use a second chart instead, i.e., 30-second or less.

As I'm working on my charting methodology again, I want to once again explore other options. I know many people seem to prefer volume charts or tick charts over time based charts. Some utilize a combination.

In the past I've done in depth studies of volume, tick and range, but I have yet found a huge advantage over time based charts. IMO, quite often they visually look the same. The exception is during times of huge volatility, but as mentioned, this can be solved by simply using a sub-minute time frame instead.

Any comments?

Which charts do you use for day trading?

PS: I seem to recall that volume/tick charts would form some nice 'bases' on swing highs and lows the last time I trialed them, but this pattern seems to be gone now.

Thanks in advance.

LaissezFaire

Reply With Quote
The following user says Thank You to LaissezFaire for this post:

Journal Challenge April 2021 results (now extended!):
Competing for $1800 in prizes from Jigsaw
looks_oneMaking a Living with the Microsby sstheo
(84 thanks from 14 posts)
looks_twoSalao's Journalby Salao
(30 thanks from 7 posts)
looks_3Deetee’s DAX Trading Journal (time based)by Deetee
(26 thanks from 9 posts)
looks_4Learning to Profit - A journey in algorithms and optionsby Syntax
(13 thanks from 8 posts)
looks_5Maybe a little bit different journalby Malykubo
(7 thanks from 5 posts)
 
 
(login for full post details)
  #3 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received


No one? Or is the topic discussed to death already...?

Reply With Quote
 
(login for full post details)
  #4 (permalink)
 noobforlyfe 
Toronto , Ontario, Canada
 
Experience: Beginner
Platform: Sierra Chart, SC
Trading: Emini ES, CL, USD/CAD
 
Posts: 424 since Jun 2016
Thanks: 1,379 given, 723 received


LaissezFaire View Post
Hi all,

My main chart for day trading (ES) is the 1-minute chart. I find that most of the time this chart gives a good summary of the price action without too much noise. At times of heightened volatility as seen recently, it can be too blunt. A solution during these periods of time may simply be to use a second chart instead, i.e., 30-second or less.

As I'm working on my charting methodology again, I want to once again explore other options. I know many people seem to prefer volume charts or tick charts over time based charts. Some utilize a combination.

In the past I've done in depth studies of volume, tick and range, but I have yet found a huge advantage over time based charts. IMO, quite often they visually look the same. The exception is during times of huge volatility, but as mentioned, this can be solved by simply using a sub-minute time frame instead.

Any comments?

Which charts do you use for day trading?

PS: I seem to recall that volume/tick charts would form some nice 'bases' on swing highs and lows the last time I trialed them, but this pattern seems to be gone now.

Thanks in advance.

LaissezFaire

I'm actually confused, how does a 30 second chart offer more clarity during increased volatility than a higher time frame chart which would be according to your charts, the 1 minute chart. When your 1 minute gets choppy and blunt, look to a higher time frame chart, for your system it could be 5 min. Someone trading 15 min chart may look at an hourly chart as their higher time frame chart.

However, you will find much more chop and noise on a 30 second or less if you use that as your HTF chart.

I use tick charts, not for any complex special reason. But because I'm used to it and i often interchange with bar charts of 2 mins and my HTF chart ranges about 5/10 minute. Price will still be displayed the same across all charts, if it moves down 5 points, however you like to see it, it will still be seen as a move down by 5 points.

Edit: Assuming we are looking at it, as it forms and not comparing 5 min bar to 1/4/24 hrs bar. one bar can represent the entire day or thousands of bars can represent an entire day. but my point is across all different charts when it does drop by 5 points. everyone sees the drop regardless of what type of chart they are using.

Some people like a 10,000 volume. Others have theirs for specific reasons, I just chose something that I like the pace of. It's not too fast and not too slow like 15 minute bar chart.

Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #5 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received


noobforlyfe View Post
I'm actually confused, how does a 30 second chart offer more clarity during increased volatility than a higher time frame chart which would be according to your charts, the 1 minute chart. When your 1 minute gets choppy and blunt, look to a higher time frame chart, for your system it could be 5 min. Someone trading 15 min chart may look at an hourly chart as their higher time frame chart.

However, you will find much more chop and noise on a 30 second or less if you use that as your HTF chart.

To me, it's the other way around. A lot can happen inside a 1-minute bar during high volatility.

Reply With Quote
The following user says Thank You to LaissezFaire for this post:
 
(login for full post details)
  #6 (permalink)
 noobforlyfe 
Toronto , Ontario, Canada
 
Experience: Beginner
Platform: Sierra Chart, SC
Trading: Emini ES, CL, USD/CAD
 
Posts: 424 since Jun 2016
Thanks: 1,379 given, 723 received


LaissezFaire View Post
To me, it's the other way around. A lot can happen inside a 1-minute bar during high volatility.

To each his own

all the best, will be following for any updates.



Sent using the futures.io mobile app

Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #7 (permalink)
 MiniP 
Market Wizard
Columbus OHIO
 
Experience: Intermediate
Platform: NinjaTrader
Broker: NinjaTrader Brokerage
Trading: ES,
 
MiniP's Avatar
 
Posts: 1,142 since May 2017
Thanks: 1,102 given, 2,898 received


noobforlyfe View Post
To each his own

all the best, will be following for any updates.



Sent using the futures.io mobile app

Id say your both right, a'lot can happen in 1 minute but most likely unless its a "black swan " event we arent going to go anywhere significant .. kinda like running on a tread mill i can turn it all the way up to 10 and it looks like I'm doing a lot but really i haven't moved an inch.


Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #8 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received

It really depends on how you use charts and what you're looking for. What if I'm a scalper?

Incidentally, I'm not, but I want low risk entries and good execution without using very wide stops. For me, the 5-minute chart is not sufficient for that purpose.

Even if you rely on higher timeframes for signals, you'd still want to drill down to a finer level if you want the best possible execution. Other people might be comfortable taking on more heat, being early, but that's not my preference.

Getting back on topic, it would be interesting to hear if people have found any particular advantages with activity based charts over time based charts...

Reply With Quote
The following user says Thank You to LaissezFaire for this post:
 
(login for full post details)
  #9 (permalink)
 iantg 
charlotte nc
 
Experience: Master
Platform: ninjatrader
Broker: NinjaTrader
Trading: Emini (ES, YM, NQ, ect.)
 
Posts: 398 since Jan 2015
Thanks: 76 given, 1,085 received

There are a few neat things you can do with tick charts that you can't do with time based charts.

You can develop an easy view of the average trade size. (In a 100 Tick bar) If you have a total volume of 1000, then you know that the average trade size in 10. You can back into this type of calculation with any size tick bar, but this will give you clues about what type of traders are in the market. Some people create trading rules around following the larger players, though I won't comment on how well this works.

If you have a really granular setting like 1 tick for example, and you run this to only update on price level changes then you can capture every price level change the market makes. Some platforms allow this, but others might not. But seeing every single price level is neat and can lead to a few edges. I won't speak to how well they work, but there are a lot of schools of thought on if a price level gets penetrated or not on the first or second pass then x,y,z will occur. At the time time if a price level gets penetrated every time, then when the market moves to this price level again... A,B,C will occur. This level of granularity may not be for most chart traders, but scalpers love this stuff, and there are some chart trading scalpers out there, just not a lot.

If you are trying to perfectly time your entry to get on the right side of the noise and capture a tick or 2, then analyzing the action at the price levels can get you on the right side of the noise often times. If you are looking too far out though (1 minute to 5 minute data) you may miss a key price level change that can throw your entry off by 1 tick.

While I will concede that it is very challenging to work with granular data as a chart trader, there are some possible benefits to it if you are fast enough. This is what playing video games as kids for 20 years prepped you for right?

Best of luck!

Ian

In the analytical world there is no such thing as art, there is only the science you know and the science you don't know. Characterizing the science you don't know as "art" is a fools game.
Visit my futures io Trade Journal Reply With Quote
The following 2 users say Thank You to iantg for this post:
 
(login for full post details)
  #10 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received


Nice insight! Thanks a lot, Ian!

Since you mentioned tick charts specifically. Do you have an opinion on volume charts also?

I know many people love volume charts, but to me, I feel that tick charts make more sense.

Curiously, I find that they usually don't look that different anyway. Usually, both tick/volume/minute look fairly similar on any given day. Unless you use really fine settings.

Reply With Quote
 
(login for full post details)
  #11 (permalink)
 MiniP 
Market Wizard
Columbus OHIO
 
Experience: Intermediate
Platform: NinjaTrader
Broker: NinjaTrader Brokerage
Trading: ES,
 
MiniP's Avatar
 
Posts: 1,142 since May 2017
Thanks: 1,102 given, 2,898 received


LaissezFaire View Post
Nice insight! Thanks a lot, Ian!

Since you mentioned tick charts specifically. Do you have an opinion on volume charts also?

I know many people love volume charts, but to me, I feel that tick charts make more sense.

Curiously, I find that they usually don't look that different anyway. Usually, both tick/volume/minute look fairly similar on any given day. Unless you use really fine settings.

take a look at a 900 tick es chart there will be tons of pull back's for entries under 2 pts or less, you can look at volume charts in the 5000-10000 range that are very similar

i prefer a 4500 tick chart but also trade a 900 tick chart

Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to MiniP for this post:
 
(login for full post details)
  #12 (permalink)
 iantg 
charlotte nc
 
Experience: Master
Platform: ninjatrader
Broker: NinjaTrader
Trading: Emini (ES, YM, NQ, ect.)
 
Posts: 398 since Jan 2015
Thanks: 76 given, 1,085 received

I don't use charts myself.... but volume charts can give you an insight into what moves the market and also how much of the volume that you see resting on DOM is speculate vs. real.

If you look at the level 2 book and you see it mostly stacked with 500 per price level, then see what a 250 volume bar would plot. Typically the transaction level holds < volume than the resting levels. If every 250 volume bar is moving 1 or more price levels when you plot it, then you can assume half of the volume on the DOM is speculative. Don't think of this as fake necessarily, but people are trying to optimize their queue positions so they cancel and resubmit.

A few insights you might pick up from this:

1. If the transacted volume is roughly 1 to 1 with the resting volume in the DOM then you are usually in a very high volatility period and this tends to indicate trending systems will perform well. In higher volatility periods, a lot of MM algo's shut down, and leave mostly stoic algos that don't cancel and resubmit as much. Also it is trickier to land cancels at the transaction level when the entire price level clears in a blip.

2. If the transacted volume is 1.5x or 1x to 2x the resting volumes in the DOM then this is probably a normal level.

3. if you see something like 1x to 4x or 1x to 8x then you are in a low volatility / ranging market. Everyone is jockeying for position in the queue and no one wants to cross the spread or get a bad fill.

3. If your tick chart is plotting roughly the same visually as your volume chart with the same # settings. Let's say 100 vs. 100, then this implies that the market is made up of smaller traders more or less. But if you see that that the volume chart at 100 = the tick chart at 1000 for example, then there are larger players in the market and the average trade size is 10. Typically the competition of this is comprised of a few large trades, let's say 100, 50, 25, and then a whole lot of 1,2,3,2,1,2,3, type of trades. But it averages out. Some people like to ride the coat tails of the big players. I won't speak to the success of this logic, but if you are looking for this type of information you can easily get it with a few steps and a little logic.

The new volumetric charts from NT 8 show more detail and break up the bid vs. ask volume. This can sometimes be useful to see if a single bar of 1000 was made up of mostly bids or asks. But there is not always a clear heuristic that can be derived from this. Often times the bid volume > the aks volume but the ask side is what breaks and the price level moves up. This has to do with how the limit order book is stacked and cancelled in relation to the market order queue. So even though it may seem like a clear way to rationalize the market, there is not always a 1 to 1 heuristic.


There is nothing like this, but in terms of charts, or information in general, what would be ideal to see would be on a per price level basis:

Starting volume
Added volume
Canceled volume
Transacted Volume

If you could see this for both the bid and ask on every price level, then you could truly rationalize the behavior of the markets. But without all of these elements there is too much guess work to infer what the root cause of a price level change was. I do a lot of data analysis with these data elements, but building this type of data set is not easy. If anyone could put these elements into a trading platform they would really have something useful IMO.

Happy Trading!

Ian



LaissezFaire View Post
Nice insight! Thanks a lot, Ian!

Since you mentioned tick charts specifically. Do you have an opinion on volume charts also?

I know many people love volume charts, but to me, I feel that tick charts make more sense.

Curiously, I find that they usually don't look that different anyway. Usually, both tick/volume/minute look fairly similar on any given day. Unless you use really fine settings.


In the analytical world there is no such thing as art, there is only the science you know and the science you don't know. Characterizing the science you don't know as "art" is a fools game.
Visit my futures io Trade Journal Reply With Quote
The following 6 users say Thank You to iantg for this post:
 
(login for full post details)
  #13 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received

Ian,

Really appreciate your comment. Very insightful and new stuff to me. I will have to give this further study and thought.


iantg View Post
I don't use charts myself.... but volume charts can give you an insight into what moves the market and also how much of the volume that you see resting on DOM is speculate vs. real.

This comment also intrigued me. What are you using then...? A quantitative or automated system?

I've done quite some work these last years myself moving away from charts, but I'm now moving back towards them to aid execution and entries.

Nice signature, by the way.

Reply With Quote
 
(login for full post details)
  #14 (permalink)
 iantg 
charlotte nc
 
Experience: Master
Platform: ninjatrader
Broker: NinjaTrader
Trading: Emini (ES, YM, NQ, ect.)
 
Posts: 398 since Jan 2015
Thanks: 76 given, 1,085 received

Thanks. I am 100% algo. I am attempting to leverage HF / MM techniques at the upper end of retail level latency as crazy as that sounds. Like many HFT systems, my system has no directional bias and is agnostic to the majority of the conventional trading mechanics most people watch.


Ian


LaissezFaire View Post
Ian,

Really appreciate your comment. Very insightful and new stuff to me. I will have to give this further study and thought.



This comment also intrigued me. What are you using then...? A quantitative or automated system?

I've done quite some work these last years myself moving away from charts, but I'm now moving back towards them to aid execution and entries.

Nice signature, by the way.


In the analytical world there is no such thing as art, there is only the science you know and the science you don't know. Characterizing the science you don't know as "art" is a fools game.
Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to iantg for this post:
 
(login for full post details)
  #15 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received

Interesting, Ian. Will be paying attention to your posts. Best of luck.

Reply With Quote
The following user says Thank You to LaissezFaire for this post:
 
(login for full post details)
  #16 (permalink)
 sptrader 
Colorado
 
Experience: Advanced
Platform: MultiCharts
Broker: IB & Iqfeed
Trading: ES , CL
 
sptrader's Avatar
 
Posts: 543 since Apr 2010
Thanks: 1,094 given, 697 received

Personally, I prefer non- time-based charts, since "time" is a variable and in trading, it's best to eliminate as many variables as possible. Volume charts are what I currently use for my main trading chart.(but I do experiment with others).
Just like mechanical trading systems, usually the fewer the number of input variables, the more stable and reliable the system is.

I'd suggest trying all of the different chart types and try to get a "feel" for the type that is the most comfortable and helps you see the markets most clearly.
Good trading

Reply With Quote
The following user says Thank You to sptrader for this post:
 
(login for full post details)
  #17 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received


sptrader View Post
Personally, I prefer non- time-based charts, since "time" is a variable and in trading, it's best to eliminate as many variables as possible. Volume charts are what I currently use for my main trading chart.(but I do experiment with others).
Just like mechanical trading systems, usually the fewer the number of input variables, the more stable and reliable the system is.

I'd suggest trying all of the different chart types and try to get a "feel" for the type that is the most comfortable and helps you see the markets most clearly.
Good trading

Personally, I pay a lot of attention to time during the trading day and I like how time charts look the same every single day. They're objective and make for rational comparison, IMO. So, for me, that's a must. Using constant settings, tick/volume charts can vary some from day to day. I'm sure many people don't mind this though.

But beyond that, I'm looking to see if I can use tick or volume charts or even range charts to improve on execution and see or read something that's hidden or less obvious on a time based chart. One simple example of this would be how a bottom could form on an activity based chart while a 1-minute bar is still printing. But I'm not convinced that this couldn't simply have been observed using a sub-minute time chart instead.

I have some old data on an older hard drive which I haven't been able to acess, but I feel very certain that volume charts a few years back had this distinct pattern around market highs and lows that I can't seem to observe now. They would often form a base, if I recall correctly. It seems like highs/lows are more sharp these days.

But I need to give it more study...

Reply With Quote
 
(login for full post details)
  #18 (permalink)
 reduktion 
Sydney, Australia
 
Experience: Beginner
Platform: multicharts
Trading: Euro Stoxx 50
 
Posts: 2 since May 2018
Thanks: 0 given, 0 received


iantg View Post
2. If the transacted volume is 1.5x or 1x to 2x the resting volumes in the DOM then this is probably a normal level.

Ian

Hi Ian,

Thanks for your comment.

I've only recently started looking at the DOM, but I found your analysis on the DOM and volume bars to be interesting.
Can you explain what you mean by resting volume and transacting volume?

Reply With Quote
 
(login for full post details)
  #19 (permalink)
 iantg 
charlotte nc
 
Experience: Master
Platform: ninjatrader
Broker: NinjaTrader
Trading: Emini (ES, YM, NQ, ect.)
 
Posts: 398 since Jan 2015
Thanks: 76 given, 1,085 received

Hi reduktion,

This type of analysis won't be very useful for 99% of most retail traders because it is looking at an extremely granular view of the market. But here is the general idea:

The picture below shows the DOM with the strong side and weak side noted:



The strong side: These are the two sides that carry resting volume to the transaction level. Because people can place orders ahead of time and wait in the queue, the strong side will typically accumulate more volume than the weak side and therefore historically the strong side beats the weak side.

The Weak Side: This represents new price levels that are spontaneously created and filled with limit orders at the birth of the price level. By Contrast this side can not pull resting orders from existing price levels into it. So this puts it at a disadvantage in the match-up against the strong side. The weak side will typically lose against the strong.

At the very center the two colored numbers represent the best bid vs. the best ask on the transaction level. For every price level, one side will pull resting volume from the strong side, and one will spontaneously build volume up, (this is the weak side) and the two will do battle until the market orders chop down the limit order volume from the side that breaks first, then the market will move to a new price level and repeat.


*** Notes below only apply to the ES:

When I speak of winning or losing between the strong side or weak side (Which side breaks first), I do not mean this anecdotally. I analyzed every price level change on the ES in 2017 and the ratio was: Strong side 65% weak side 35%. In times of higher volatility this becomes more even and you will see anywhere from a 50% / 50% split to as much as 45% strong side to 55% weak side. I think you will have hard time finding days where the weak side ever beats the strong side by more than 5% to 10%. In medium volatility it is usually 55% to 60% strong side winning. In low volatility this is usually 65% to 70% strong side winning.

All instruments will likely have a similar pattern, but the ones that are more mechanical will be more like the ES. I also looked the the NQ, and YM and they favored the strong side less, but this is obvious because they have less resting volumes typically and are not as automated by trading algo's.

Regarding transacted volume vs. resting volume:

Resting volume just means the volumes you see posted on the DOM from the strong side on each price level. In this example (Enclosed image) We can see that outside of the transaction level, the lowest volume on both the bid and ask is at 300 ish, and this goes as high as 800 to 900.

Transacted Volume: The transacted volume refers to how many contracts traded at the bid or ask on a given price level. *** This is the part where this analysis will lose most retail traders. Most traders aggregate their charting tools up based on fairly arbitrary units of measure such as (By # of trades, (Ticks), by Time, (Every 10 seconds, every 1 minute) etc. The action that occurs on an individual price level is important and should ultimately be viewed for each price level to make heads of tails of what occurred. Even though I don't use charts, I think the best unit of measure to see is the price levels. I would only look at this if I were a chart trader.



Here is an example of a series of events that occur on a price level:

The Starting volume from the strong side = 400
The Starting volume from the weak side = 0.
As soon as the price level is created and traders get this information the following occurs:

1. 100 HF traders immediately cancel the strong side volume, moving it from 400 to 300.
2. HF traders add 200 limit orders to the weak side. (So now we have 300 SS vs. 200 WS)

*** 1 and 2 above occur before any market orders even hit the price level.

3. Now 100 market orders hit the strong side and chop down the volume from 300 to 200
4. Market orders do not seem as interest in the weak side and only take out 20 of the 200.

*** Reaction from points 3 and 4.

5. Based on points 3 and 4 above: HF traders fill up even more volume on the weak side, they are now betting this side will win because of light interest from the first round of market orders. So now the weak side is up to 300
6. HF traders that were on the strong side see the weak side volume increasing and the market orders to their side keep coming, so it is becoming obvious that this is losing battle. Out of the 150 or so that are left. 100 cancel strategically getting out of the way, avoiding the impending doom.

7. The Ending: There are now 50 or so unlucky traders still left on the strong side that did not land their cancel in time, or did not know to cancel (Retail traders) The market orders quickly take them out and this side breaks, the strong side loses and the weak side wins.

From the points 1- 7. We can observe that around 50% of the starting volume from the strong side was strategically canceled and not transacted. Reconciling this, you can simply get an output of the transacted volume by Bid and Ask for this price level and see it.

You take the starting volume before the price level the transaction level:

1. 400
2. The transacted volume: In this example, 200 or so
3. So the delta here is 400 - 200. So this implies that 200 of the original volume was speculative, strategically cancelled and this can give you clues about a whole host of other things, but most of this will be useless to retail traders.


I think this will give you some ideas:

The edge case here is not really something that most people would ever pick up on or use. But comparing transacted volume to resting volume can give us clues about the market dynamics. You can also gain valuable insights analyzing how many orders were submitted but went unfilled on a given price level, especially on the side that won.

Hopefully this fields you query, and peaks your interest in micro structures a little.

Happy Trading

Ian







reduktion View Post
Hi Ian,

Thanks for your comment.

I've only recently started looking at the DOM, but I found your analysis on the DOM and volume bars to be interesting.
Can you explain what you mean by resting volume and transacting volume?


In the analytical world there is no such thing as art, there is only the science you know and the science you don't know. Characterizing the science you don't know as "art" is a fools game.
Visit my futures io Trade Journal Reply With Quote
The following 2 users say Thank You to iantg for this post:
 
(login for full post details)
  #20 (permalink)
 MiniP 
Market Wizard
Columbus OHIO
 
Experience: Intermediate
Platform: NinjaTrader
Broker: NinjaTrader Brokerage
Trading: ES,
 
MiniP's Avatar
 
Posts: 1,142 since May 2017
Thanks: 1,102 given, 2,898 received


iantg View Post
Hi reduktion,

This type of analysis won't be very useful for 99% of most retail traders because it is looking at an extremely granular view of the market. But here is the general idea:

The picture below shows the DOM with the strong side and weak side noted:



The strong side: These are the two sides that carry resting volume to the transaction level. Because people can place orders ahead of time and wait in the queue, the strong side will typically accumulate more volume than the weak side and therefore historically the strong side beats the weak side.

The Weak Side: This represents new price levels that are spontaneously created and filled with limit orders at the birth of the price level. By Contrast this side can not pull resting orders from existing price levels into it. So this puts it at a disadvantage in the match-up against the strong side. The weak side will typically lose against the strong.

At the very center the two colored numbers represent the best bid vs. the best ask on the transaction level. For every price level, one side will pull resting volume from the strong side, and one will spontaneously build volume up, (this is the weak side) and the two will do battle until the market orders chop down the limit order volume from the side that breaks first, then the market will move to a new price level and repeat.


*** Notes below only apply to the ES:

When I speak of winning or losing between the strong side or weak side (Which side breaks first), I do not mean this anecdotally. I analyzed every price level change on the ES in 2017 and the ratio was: Strong side 65% weak side 35%. In times of higher volatility this becomes more even and you will see anywhere from a 50% / 50% split to as much as 45% strong side to 55% weak side. I think you will have hard time finding days where the weak side ever beats the strong side by more than 5% to 10%. In medium volatility it is usually 55% to 60% strong side winning. In low volatility this is usually 65% to 70% strong side winning.

All instruments will likely have a similar pattern, but the ones that are more mechanical will be more like the ES. I also looked the the NQ, and YM and they favored the strong side less, but this is obvious because they have less resting volumes typically and are not as automated by trading algo's.

Regarding transacted volume vs. resting volume:

Resting volume just means the volumes you see posted on the DOM from the strong side on each price level. In this example (Enclosed image) We can see that outside of the transaction level, the lowest volume on both the bid and ask is at 300 ish, and this goes as high as 800 to 900.

Transacted Volume: The transacted volume refers to how many contracts traced at the bid or ask on a given price level. *** This is the part where this analysis will lose most retail traders. Most traders aggregate their charting tools up based on fairly arbitrary units of measure such as (By # of trades, (Ticks), by Time, (Every 10 seconds, every 1 minute) etc. The action that occurs on an individual price level is important and should ultimately be viewed for each price level to make heads of tails of what occurred.

For example:

The Starting volume from the strong side = 400
The Starting volume from the weak side = 0.
As soon as the price level is created and traders get this information the following occurs:

1. 100 HF traders immediately cancel the strong side volume, moving it from 400 to 300.
2. HF traders add 200 limit orders to the weak side. (So now we have 300 SS vs. 200 WS)

*** 1 and 2 above occur before any market orders even hit the price level.

3. Now 100 market orders hit the strong side and chop down the volume from 300 to 200
4. Market orders do not seem as interest in the weak side and only take out 20 of the 200.

*** Reaction from points 3 and 4.

5. Based on points 3 and 4 above: HF traders fill up even more volume on the weak side, they are now betting this side will win because of light interest from the first round of market orders. So now the weak side is up to 300
6. HF traders that were on the strong side see the weak side volume increasing and the market orders to their side keep coming, so it is becoming obvious that this is losing battle. Out of the 150 or so that are left. 100 cancel strategically getting out of the way, avoiding the impending doom.

7. The Ending: There are now 50 or so unlucky traders still left on the strong side that did not land their cancel in time, or did not know to cancel (Retail traders) The market orders quickly take them out and this side breaks, the strong side loses and the weak side wins.

From the points 1- 7. We can observe that around 50% of the starting volume from the strong side was strategically canceled and not transacted. Reconciling this, you can simply get an output of the transacted volume by Bid and Ask for this price level and see it.

You take the starting volume before the price level the transaction level:

1. 400
2. The transacted volume: In this example, 200 or so
3. So the delta here is 400 - 200. So this implies that 200 of the original volume was speculative, strategically cancelled and this can give you clues about a whole host of other things, but most of this will be useless to retail traders.


I think this will give you some ideas:

The edge case here is not really something that most people would ever pick up on or use. But comparing transacted volume to resting volume can give us clues about the market dynamics.

Hopefully this fields you query, and peaks your interest in micro structures a little.

Happy Trading

Ian

This is EXTREMELY interesting to me, honestly I understand about 85% of what you said above the last few points kind of ran together, any chance you have some recommended material/starting point for understating this?

Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #21 (permalink)
 iantg 
charlotte nc
 
Experience: Master
Platform: ninjatrader
Broker: NinjaTrader
Trading: Emini (ES, YM, NQ, ect.)
 
Posts: 398 since Jan 2015
Thanks: 76 given, 1,085 received

Hi MiniP,

Thanks for expressing interest in this area...

I am not aware of any material that covers the topic of price levels clearing and provides statistics / historical information.
Everything I have learned on subject I have had to learn it the hard way by trading, collecting the right meta-data, then analyzing it later. (Plowing through tens of million rows of level 2 data in SQL, etc.) I suppose what we are really talking about here are the statistics underlying micro-structure. There are plenty of books on micro-structures generically but none that I am aware of that share the statistical outcomes. Here is what I am referring to, and this will get you thinking the right way about it.

We all know that if you pick the wrong (bid / ask) side to place your limit order, you will get filled. You will always catch a 100% fill rate if you are on the wrong side. (Volume goes to 0, the price level breaks. Since your order was for at least 1 contract, you had to get filled for it to go to 0)

What we don't know obviously, and what a good deal of my research is based on now days, is around optimizing for how far up in the queue you need to be to get filled on the side that actually wins: So if the Bid goes to 0 first and breaks, then how many people got filled from the Ask queue. The top 10%, the top 50%, the top 75%? What are some historical data points across different instruments, markets, volatility levels, etc.

What most people assume are "good signals" in the market are actually causing them more harm then good, and here is why. Your fill rate on the losing side will be 100%, but your fill rate on the winning side will be extremely low if you get there late. A good bit of order flow traders get killed by reading the transaction level.... seeing that one side has 2x or 3x the volume of the other and then jumping on the side that has the higher volume. This seems like a safe bet, they are picking the side that will win surely. But in practice, because they are late, the only fills they will get, will be toxic fills. So even if there is an edge to seeing obvious volume ratios on the transaction level, acting on this will play out like this:

1. Fill Rate From being on the side that loses: 100%
2. Fill Rate from being on the side that wins: Very Very low because you are late. Maybe 10% or < depending on how late you decide to join the party.

So if your "so called edge" I.E (the bid volume is 3 X the Ask volume.... Join the bid queue) actually leads to the market moving higher by 1 tick on the next price level 70% of the time, and it is wrong 30% of the time (I am being generous here with the edge ratio). Here is how it will play out for you.

Full Expectancy: over 100 trades
1. 30% of the time your bet is wrong, but because your bet was wrong, everyone in the bid queue including you got a toxic fill: So out of 100 trades, this = 30 trades that were losers, all of which you got filled on.
2. 70% of the time your bet was right, the bid side won! But, you were late to the party and only got filled 10% of the time. so 100 trades * 70% bet was right * 10% order was filled = 7 trades you won on.

So you will end up with 30 losers and 7 winners. So that is how the full expectancy of this so called "good bet" works in practice. And everyone is doing it just like this and they are all getting killed.

The first thing I suppose you should research are toxic fills, and understand how these are impacting your trading. This is one of the least obvious, but most important aspects of trading that I think everyone overlooks, or just doesn't understand. But I just gave a high level explanation of it, so hopefully it helps.

I haven't seen any threads here that cover these types of topics in detail, so I may start one if there interest and cover this type of thing in more detail.

Happy Trading!

Ian






MiniP View Post
This is EXTREMELY interesting to me, honestly I understand about 85% of what you said above the last few points kind of ran together, any chance you have some recommended material/starting point for understating this?


In the analytical world there is no such thing as art, there is only the science you know and the science you don't know. Characterizing the science you don't know as "art" is a fools game.
Visit my futures io Trade Journal Reply With Quote
The following 5 users say Thank You to iantg for this post:
 
(login for full post details)
  #22 (permalink)
 lax99 
Denver
 
Experience: Intermediate
Platform: Bookmap and Jigsaw DOM
Broker: Stage 5 Trading
Trading: ZN
 
lax99's Avatar
 
Posts: 433 since Jun 2015
Thanks: 623 given, 813 received

I think it's quite a bit easier to see the concepts which ian speaks of in a thicker market. I don't know about the statistical bid/offer stuff. Large players can bait either side with ease and you need to understand the context of the moment to be able to reliably read it.

Toxic fills, however, are the absolutely bane of my trading existence. I trade ZN which is very liquid and often slow. I use a 2 tick stop and I usually look to make 3 or 4 ticks, for example. When I'm bid/offered at some price and the market sweeps through me and I'm immediately looking at a -1, this is usually my sign that I'm already dead in the water and that I made the wrong call. Again, it's situational and it depends on the read, but if I'm passively looking to interact with the market, I never want the market to tick against me. I want to be filled long and have 3000 bid behind me which I can hit into if something changes.

Reply With Quote
The following 3 users say Thank You to lax99 for this post:
 
(login for full post details)
  #23 (permalink)
 MiniP 
Market Wizard
Columbus OHIO
 
Experience: Intermediate
Platform: NinjaTrader
Broker: NinjaTrader Brokerage
Trading: ES,
 
MiniP's Avatar
 
Posts: 1,142 since May 2017
Thanks: 1,102 given, 2,898 received


iantg View Post
Hi MiniP,

Thanks for expressing interest in this area...

I am not aware of any material that covers the topic of price levels clearing and provides statistics / historical information.
Everything I have learned on subject I have had to learn it the hard way by trading, collecting the right meta-data, then analyzing it later. (Plowing through tens of million rows of level 2 data in SQL, etc.) I suppose what we are really talking about here are the statistics underlying micro-structure. There are plenty of books on micro-structures generically but none that I am aware of that share the statistical outcomes. Here is what I am referring to, and this will get you thinking the right way about it.

We all know that if you pick the wrong (bid / ask) side to place your limit order, you will get filled. You will always catch a 100% fill rate if you are on the wrong side. (Volume goes to 0, the price level breaks. Since your order was for at least 1 contract, you had to get filled for it to go to 0)

What we don't know obviously, and what a good deal of my research is based on now days, is around optimizing for how far up in the queue you need to be to get filled on the side that actually wins: So if the Bid goes to 0 first and breaks, then how many people got filled from the Ask queue. The top 10%, the top 50%, the top 75%? What are some historical data points across different instruments, markets, volatility levels, etc.

What most people assume are "good signals" in the market are actually causing them more harm then good, and here is why. Your fill rate on the losing side will be 100%, but your fill rate on the winning side will be extremely low if you get there late. A good bit of order flow traders get killed by reading the transaction level.... seeing that one side has 2x or 3x the volume of the other and then jumping on the side that has the higher volume. This seems like a safe bet, they are picking the side that will win surely. But in practice, because they are late, the only fills they will get, will be toxic fills. So even if there is an edge to seeing obvious volume ratios on the transaction level, acting on this will play out like this:

1. Fill Rate From being on the side that loses: 100%
2. Fill Rate from being on the side that wins: Very Very low because you are late. Maybe 10% or < depending on how late you decide to join the party.

So if your "so called edge" I.E (the bid volume is 3 X the Ask volume.... Join the bid queue) actually leads to the market moving higher by 1 tick on the next price level 70% of the time, and it is wrong 30% of the time (I am being generous here with the edge ratio). Here is how it will play out for you.

Full Expectancy: over 100 trades
1. 30% of the time your bet is wrong, but because your bet was wrong, everyone in the bid queue including you got a toxic fill: So out of 100 trades, this = 30 trades that were losers, all of which you got filled on.
2. 70% of the time your bet was right, the bid side won! But, you were late to the party and only got filled 10% of the time. so 100 trades * 70% bet was right * 10% order was filled = 7 trades you won on.

So you will end up with 30 losers and 7 winners. So that is how the full expectancy of this so called "good bet" works in practice. And everyone is doing it just like this and they are all getting killed.

The first thing I suppose you should research are toxic fills, and understand how these are impacting your trading. This is one of the least obvious, but most important aspects of trading that I think everyone overlooks, or just doesn't understand. But I just gave a high level explanation of it, so hopefully it helps.

I haven't seen any threads here that cover these types of topics in detail, so I may start one if there interest and cover this type of thing in more detail.

Happy Trading!

Ian

Ian,

I really think you should start a thread on this, what your saying makes perfect sense especially the not getting filled on winning trades.

Few questions i apologize if there are ignorant i don't trade remotely close to this way:
1) how do you see where in the queue you currently are?
2) bascially alot of what you are saying depends on the resting lvl of volume and where in line you fall?

Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #24 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received

Good stuff, guys.

Related to this topic:

Do you guys have any preference with regards to what bars you use to display your charts?

Candlesticks? Regular bars? HiLo bars?

For my time based charts I have a clear preference for candlesticks. I like to observe wicks and consider large wicks/tails to indicate exhaustion at price levels. This is interesting.

But for faster charts, I might be interested in compressing my chart a bit and the wicks might be less interesting since there will be a 'lot' of bars anyway.

Reply With Quote
 
(login for full post details)
  #25 (permalink)
 MiniP 
Market Wizard
Columbus OHIO
 
Experience: Intermediate
Platform: NinjaTrader
Broker: NinjaTrader Brokerage
Trading: ES,
 
MiniP's Avatar
 
Posts: 1,142 since May 2017
Thanks: 1,102 given, 2,898 received


LaissezFaire View Post
Good stuff, guys.

Related to this topic:

Do you guys have any preference with regards to what bars you use to display your charts?

Candlesticks? Regular bars? HiLo bars?

For my time based charts I have a clear preference for candlesticks. I like to observe wicks and consider large wicks/tails to indicate exhaustion at price levels. This is interesting.

But for faster charts, I might be interested in compressing my chart a bit and the wicks might be less interesting since there will be a 'lot' of bars anyway.

I use regular candle sticks, like you said i also like to see wicks and rejection areas.. ive been playing around with a 4500 tick chart with a line on close with a swing lvl indicator to show ticks and this is a very clean way to trade if you like P/A.

Will be testing this in the upcoming week.

Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to MiniP for this post:
 
(login for full post details)
  #26 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received


MiniP View Post
I use regular candle sticks, like you said i also like to see wicks and rejection areas.. ive been playing around with a 4500 tick chart with a line on close with a swing lvl indicator to show ticks and this is a very clean way to trade if you like P/A.

Will be testing this in the upcoming week.

Cool.

Would you mind sharing a screen if this isn't 'secret'?

Reply With Quote
 
(login for full post details)
  #27 (permalink)
 MiniP 
Market Wizard
Columbus OHIO
 
Experience: Intermediate
Platform: NinjaTrader
Broker: NinjaTrader Brokerage
Trading: ES,
 
MiniP's Avatar
 
Posts: 1,142 since May 2017
Thanks: 1,102 given, 2,898 received


LaissezFaire View Post
Cool.

Would you mind sharing a screen if this isn't 'secret'?

Here is a look at yesterdays chart, it you see a floating set of dots in a line that is where the wicks end.

Still working on entries and exits but this seems to filter out some of the fluff

Attached Thumbnails
Click image for larger version

Name:	fiooo.JPG
Views:	98
Size:	124.7 KB
ID:	249554  
Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to MiniP for this post:
 
(login for full post details)
  #28 (permalink)
 cory 
the coin hunter
virginia
 
Experience: Intermediate
Platform: ninja
Trading: NQ
 
cory's Avatar
 
Posts: 6,042 since Jun 2009
Thanks: 854 given, 7,910 received


MiniP View Post
I ... with a line on close....

set it to bar HL instead then you won't have floating dots

Reply With Quote
The following user says Thank You to cory for this post:
 
(login for full post details)
  #29 (permalink)
 MiniP 
Market Wizard
Columbus OHIO
 
Experience: Intermediate
Platform: NinjaTrader
Broker: NinjaTrader Brokerage
Trading: ES,
 
MiniP's Avatar
 
Posts: 1,142 since May 2017
Thanks: 1,102 given, 2,898 received


cory View Post
set it to bar HL instead then you won't have floating dots

Then it would be different then the regular 4500 candle stick chart, right now this lines up with my normal chart but I'll test it out and see how I like it

Sent using the futures.io mobile app

"Truth is not what you want it to be; it is what it is, and you must bend to its power or live a lie"-Miyamoto Musashi
Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to MiniP for this post:
 
(login for full post details)
  #30 (permalink)
LaissezFaire
Oslo + Norway
 
 
Posts: 185 since May 2016
Thanks: 182 given, 81 received

Looking at the daily high on the 14th of May using 1-minute, 30 second, 50 tick and 250 volume:



It basically looks the same to me.

There might be times where the difference is more obvious of course, but generally, I find that these type of charts end up looking fairly similar.

Reply With Quote
The following user says Thank You to LaissezFaire for this post:
 
(login for full post details)
  #31 (permalink)
 cory 
the coin hunter
virginia
 
Experience: Intermediate
Platform: ninja
Trading: NQ
 
cory's Avatar
 
Posts: 6,042 since Jun 2009
Thanks: 854 given, 7,910 received


LaissezFaire View Post
Looking at the daily high on the 14th of May using 1-minute, 30 second, 50 tick and 250 volume:

,,,

It basically looks the same to me.

There might be times where the difference is more obvious of course, but generally, I find that these type of charts end up looking fairly similar.

they key thing is to filter enough noise to see paterns/levels clearer. I use 2m for better illustration 1m would be too much noise.

Attached Thumbnails
Click image for larger version

Name:	Capture.PNG
Views:	98
Size:	183.7 KB
ID:	249865  
Reply With Quote
The following 4 users say Thank You to cory for this post:
 
(login for full post details)
  #32 (permalink)
 HileTroy 
Colorado Springs
 
Experience: Intermediate
Platform: Track n Trade
Trading: Currency
 
Posts: 1 since May 2018
Thanks: 0 given, 1 received

I use range 6 range 12 all the time

Reply With Quote
The following user says Thank You to HileTroy for this post:


futures io Trading Community Traders Hideout > Anyone find any edge with activity based charts over time based charts?


Last Updated on May 20, 2018


Upcoming Webinars and Events
 

NinjaTrader Indicator Challenge!

Ongoing
 

Journal Challenge w/$1,800 in prizes!

April

Seven Trading Mistakes Solved With Smart Trading Tools w/Brannigan Barrett

Elite only
     



Copyright © 2021 by futures io, s.a., Av Ricardo J. Alfaro, Century Tower, Panama, +507 833-9432, info@futures.io
All information is for educational use only and is not investment advice.
There is a substantial risk of loss in trading commodity futures, stocks, options and foreign exchange products. Past performance is not indicative of future results.
no new posts